CN104813184B - Calibration to the dynamic digital imaging system for detecting the defects of production stream - Google Patents

Calibration to the dynamic digital imaging system for detecting the defects of production stream Download PDF

Info

Publication number
CN104813184B
CN104813184B CN201380061353.6A CN201380061353A CN104813184B CN 104813184 B CN104813184 B CN 104813184B CN 201380061353 A CN201380061353 A CN 201380061353A CN 104813184 B CN104813184 B CN 104813184B
Authority
CN
China
Prior art keywords
system
color
calibration
platform
unit
Prior art date
Application number
CN201380061353.6A
Other languages
Chinese (zh)
Other versions
CN104813184A (en
Inventor
里克·温德尔·巴耶玛
威尔弗雷德·马塞尔林·鲍瑞格
斯科特·费根
桑蔡·兰格
克尔温·布兰得利
大卫·雷·沃伦
Original Assignee
福瑞托-雷北美有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority to US201261729883P priority Critical
Priority to US61/729,883 priority
Priority to US13/803,484 priority
Priority to US13/803,484 priority patent/US9699447B2/en
Application filed by 福瑞托-雷北美有限公司 filed Critical 福瑞托-雷北美有限公司
Priority to PCT/US2013/071490 priority patent/WO2014082010A1/en
Publication of CN104813184A publication Critical patent/CN104813184A/en
Application granted granted Critical
Publication of CN104813184B publication Critical patent/CN104813184B/en

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N17/00Diagnosis, testing or measuring for television systems or their details
    • H04N17/002Diagnosis, testing or measuring for television systems or their details for television cameras
    • BPERFORMING OPERATIONS; TRANSPORTING
    • B07SEPARATING SOLIDS FROM SOLIDS; SORTING
    • B07CPOSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
    • B07C5/00Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
    • B07C5/34Sorting according to other particular properties
    • B07C5/342Sorting according to other particular properties according to optical properties, e.g. colour
    • B07C5/3422Sorting according to other particular properties according to optical properties, e.g. colour using video scanning devices, e.g. TV-cameras
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01JMEASUREMENT OF INTENSITY, VELOCITY, SPECTRAL CONTENT, POLARISATION, PHASE OR PULSE CHARACTERISTICS OF INFRA-RED, VISIBLE OR ULTRA-VIOLET LIGHT; COLORIMETRY; RADIATION PYROMETRY
    • G01J3/00Spectrometry; Spectrophotometry; Monochromators; Measuring colours
    • G01J3/28Investigating the spectrum
    • G01J3/2803Investigating the spectrum using photoelectric array detector
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01JMEASUREMENT OF INTENSITY, VELOCITY, SPECTRAL CONTENT, POLARISATION, PHASE OR PULSE CHARACTERISTICS OF INFRA-RED, VISIBLE OR ULTRA-VIOLET LIGHT; COLORIMETRY; RADIATION PYROMETRY
    • G01J3/00Spectrometry; Spectrophotometry; Monochromators; Measuring colours
    • G01J3/46Measurement of colour; Colour measuring devices, e.g. colorimeters
    • G01J3/50Measurement of colour; Colour measuring devices, e.g. colorimeters using electric radiation detectors
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01JMEASUREMENT OF INTENSITY, VELOCITY, SPECTRAL CONTENT, POLARISATION, PHASE OR PULSE CHARACTERISTICS OF INFRA-RED, VISIBLE OR ULTRA-VIOLET LIGHT; COLORIMETRY; RADIATION PYROMETRY
    • G01J3/00Spectrometry; Spectrophotometry; Monochromators; Measuring colours
    • G01J3/46Measurement of colour; Colour measuring devices, e.g. colorimeters
    • G01J3/52Measurement of colour; Colour measuring devices, e.g. colorimeters using colour charts
    • G01J3/524Calibration of colorimeters
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01NINVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00Investigating or analysing materials by the use of optical means, i.e. using infra-red, visible or ultra-violet light
    • G01N21/84Systems specially adapted for particular applications
    • G01N21/88Investigating the presence of flaws or contamination
    • G01N21/93Detection standards; Calibrating baseline adjustment, drift correction

Abstract

A kind of calibration system and method for dynamic digital imaging system, the dynamic digital imaging system are used to detect the defects of product stream of motion.The calibration system has a scaoffold above supply unit, for receiving the ceramic tile with reference color.Scaoffold can make product to be tested on the supply unit surface of lower section by make it that calibration and recalibration procedure can be carried out continuously during image obtains, without interrupting product stream.

Description

Calibration to the dynamic digital imaging system for detecting the defects of production stream

Technical field

In general, the present invention relates to digital image analysis, more specifically, it is related to one kind to for detecting continuous motion Production stream digital picture in exception or defect dynamic digital imaging system calibration.

Background technology

Exception or defects detection are carried out in multiple fields using graphical analysis.Specifically, food processing row Already through digital image analysis is used in combination with the continuous conveyor in automated food screening system.For example, some are existing Method in technology can check food and other quality with visual confirmable product or article.As any advanced Imaging or spectrum analysis are the same, and detected image is needed compared with a reference (reference), it means that is System needs to calibrate, to ensure the uniformity of measurement.But in conventional images analytical technology required for known system and method Calibration steps will often interrupt the image acquisition for defects detection.This defect detecting system and tinuous production (for example, Convey the continuous conveying system of the product of defect to be detected) with the use of when can have problem.

It is therefore desirable to have one kind need not make production induction system slow down or shut down that digital defect detecting system can be entered Row calibration and the system and method for recalibration.In addition, improved digital defect detecting system is required to existing Product delivery system is reequiped.Finally, it is also necessary to have and IMAQ and analysis system are provided accurately during continuous production The digital defect detecting system and method calibrated in real time

Brief summary of the invention

According to an aspect of the invention, there is provided a kind of dynamic digital to for detecting the defects of production stream is imaged The method that system is calibrated, this method can substantially eliminate or reduce the shortcomings that related to static calibration method before.One In individual embodiment, this method comprises the step of:(a) placed on the platform being increased to above conveyer in camera field of view One block of calibration color ceramic tile;(b) the light absorbs data of the calibration color are obtained;(c) the light absorbs data of the calibration color are calculated Composition color intensity;(d) the composition color intensity of the calibration color is normalized;(e) platform is removed into video camera Visual field;(f) the light absorbs data of product are obtained;(g) the composition color intensity of the light absorbs data of these products is calculated;(h) it is right The composition color intensity of these products is normalized;And (i) is strong according to the composition color of the calibration color and these products Degree, calculate the performance figure score value of these products.

According to another aspect of the present invention, a kind of be used for the dynamic digital for detecting the defects of production stream is disclosed The calibration system that imaging system is calibrated.In certain embodiments, the calibration system includes:Supply unit, it has an edge First framework of at least a portion width extension of supply unit;Platform, it is single that it is promoted to conveying at least one bearing The top of member, the bearing is movably coupled to the track in the first framework;And the slit in the platform, for connecing Receive the color tile for including multiple reference color sections.Some embodiments also include being coupled to directly across the width of supply unit Second framework of the first framework;Imaging unit, it includes video camera, camera window and control/data wire with visual field; And/or control unit, it includes processor, memory and display device, and the wherein imaging unit also passes through the control/data wire Communicated to connect with the control unit.

Certain embodiments of the present invention can provide multiple technologies advantage.For example, according to one embodiment of present invention, logarithm The calibration of word imaging system can substantially be carried out simultaneously with motion of the object along induction system.And implement with one of the present invention Another related technical advantage of example is that it serves many purposes.Feature provided by the present invention can be applied to quality guarantor Any induction system of the material production of card.Although embodiment as described herein employs the example of food transport system, this System and method in disclosure content can be easily adapted to any continuous production environment.The industry of disclosed method and system Include but is not limited to the example of application type:Sweets, salty snacks, candy are bakeed (for example, jelly beans, chocolate, sugar-coat are skilful Gram force, taffy and colored fruity snacks), woodwork, paper products, textile, and other many fields.

Certain embodiments of the present invention can have part or all of above-mentioned advantage.Pass through the following drawings, explanation and power Profit requires that people in the art person will readily appreciate that other technologies advantage.

Brief description

In order to which the present invention and its advantage is more fully appreciated, following explanation and accompanying drawing are referred to, wherein:

Fig. 1 is shown according to a specific embodiment, and system is imaged to the dynamic digital for detecting the defects of production stream The perspective view for the calibration system calibrated of uniting.

Fig. 2 is shown according to a specific embodiment, and system is imaged to the dynamic digital for detecting the defects of production stream The perspective view for the calibration system calibrated of uniting.

Fig. 3 is shown according to a specific embodiment, and system is imaged to the dynamic digital for detecting the defects of production stream The perspective view for the calibration system calibrated of uniting.

Fig. 4 is shown according to a specific embodiment, and system is imaged to the dynamic digital for detecting the defects of production stream The flow chart for the calibration method that system is calibrated.

Fig. 5 A and 5B are shown according to one embodiment, the respectively calibration at conveyer belt height and a height The demonstration row scanning of reference unit.

It is respectively to scan initial data according to the demonstration row of one embodiment and normalize to scheme shown in Fig. 6 A and 6B.

Fig. 7 is shown according to an alternate embodiments, and the dynamic digital for detecting the defects of production stream is imaged The perspective view for the calibration system that system is calibrated.

Fig. 8 is shown according to an alternate embodiments, and the dynamic digital for detecting the defects of production stream is imaged The top view for the calibration system that system is calibrated.

Describe in detail

Fig. 1 shows the dynamic digital imaging system for being used to detect the defects of production stream according to some embodiments. Herein, " dynamic digital imaging system " this term refers to any acquisition continuously moving object that can be operable to (for example, even Continue mobile production stream) digital picture system.System 100 includes a control unit, a supply unit, a calibration Reference unit, a lighting unit and an imaging unit.Each unit is detailed further below.

In at least one embodiment, control unit, which is one, has processor, memory, display device (for example, monitoring Device) and input equipment (for example, keyboard, mouse or touch pad) computer.In certain embodiments, control unit is also equipped with net Network interface, and it is connected to network.One advantage of this embodiment is system 100 can be controlled in remote location. In certain embodiments, control unit control supply unit, calibration reference unit, lighting unit 160 and imaging unit.At other In embodiment, supply unit is individually controlled by operator or another system.As detailed further below, calibration reference unit also may be used Controlled manually by operator.In certain embodiments, lighting unit 160 and imaging unit are carried out by the autonomous system being in communication with each other Control.And in another embodiment, manually adjust the intensity of light source 162 according to the data from acquired image by operator. In an alternative embodiment, lighting unit 160 is a static cell, and control unit is adjusted by way of application percentage factor Illumination intensity data.

In certain embodiments, supply unit is conveyer belt 102.In other embodiments, supply unit is to support With any system for conveying continuous product stream.In certain embodiments, imaging unit, lighting unit and reference color unit are set to It can be moved to reach in one from conveyer belt 102 with production stream in the portable sandwich type element (as shown in Figure 1) of another location.This is portable Sandwich type element possesses the advantage that easily can be reequiped to existing induction system.

In one embodiment shown in Fig. 1, calibration reference unit includes platform 104, framework 106, track 108, pillar 110th, along the slit 112 of platform 104, and one or more color tile 120.It should be noted that although Fig. 1 implementation exemplifies Two pillars 110, but in one embodiment, calibration reference unit only includes a pillar 110.Thus, it is however generally that, school Quasi- reference unit includes at least one pillar 110.In fig. 2 so that closer in one embodiment shown in distance, platform 104 is by band The framework 106 for having track 108 supports, and pillar 110 is freely slidable in track 108.In such an embodiment, calibration is with reference to single Member can move along perpendicular or parallel to the direction of the moving direction of conveyer belt 102.Thus enable calibration reference unit according to need It is moved into the visual field of imaging unit or is removed from the visual field of imaging unit.For example, when system 100 is not in calibration procedure When, the pillar 110 of reference unit is calibrated with the direction parallel to conveying Tape movement, and the visual field of imaging unit is skidded off along track 108, Imaging unit is set to obtain the image of the product stream on conveyer belt 102.In turn, when system 100 is in calibration or recalibration During pattern, calibration reference unit slips into the visual field of imaging unit.In certain embodiments, the operator of system 100 is with side manually Formula will calibrate reference unit into and out visual field, and in other embodiments, and control unit is in calibrating die according to system 100 Formula or detection pattern, are moved to it automatically.Although fig 1 illustrate that the simple pillar 110 slided along track 108, but can Substituted using reference unit can will be calibrated into and out any mechanical device of the visual field of imaging unit.Some examples Including roller, caster, wheel, magnetic orbit or increasingly complex robot control piece.

In general, platform 104 there can be the height of a definable, at the height, one or more can be passed through Platform is increased to the top of conveyer belt 102 by individual pillar 110 or suitable substitute.The height of pillar 110 can adjust, such as It is adjusted according to the size and dimension of the product conveyed on conveyer belt 102.In one embodiment, the height of pillar 110 It is sufficiently high so that between the surface of conveyer belt 102 and the bottom of platform 104 at least 3 inches (about 7 to 8cm) gap.Its is favourable Part is, system 100 can be calibrated, without interrupting the continuous flowing of product stream.

Platform 104 has a slit 112, and one or more color tile 120 can be slided and supported in it.It is color Color ceramic tile 120 can be the monolithic ceramic tile for having a kind of color, have the monolithic ceramic tile of multiple color, or have a variety of face The polylith ceramic tile of color.Color tile 120 is made up of any suitable material, including paper, polymer (including superhigh molecular weight polyethylene Alkene, acrylonitrile-butadiene-styrene copolymer (ABS), polyformaldehyde, nylon, polytetrafluoroethylene (PTFE) etc.), metal, ceramics, composite wood Material or timber etc..The size of color tile 120 can also change according to the demand of imaging unit.In one embodiment, it is colored The width of ceramic tile 120 is identical with the width of conveyer belt 102, and the product stream on conveyer belt 102 is placed.Shown in Fig. 1 Specific embodiment in, about 48 inches of the width (about 122cm) of color tile 120, length between 6 to 8 inches (about 15 to 20cm).In one embodiment, when a color tile is slided perpendicular to conveyer belt direct of travel, color tile can be with It is relatively small, because single pixel size is about 0.023 square inch.Light to prevent and reflecting from peripheral region occurs dry Relate to, preferably use less color tile in one embodiment.Another advantage using smaller color tile is being capable of spirit Smaller and lighter supporting device is used livingly.In certain embodiments, the size range of color tile from 0.1 square inch to 0.25 square inch or so.It should be noted that during using less ceramic tile, supporting device should be almost for imaging unit " invisible ", because it is possible to influence the local lighting in ceramic tile region.In another embodiment, one piece of narrower width Color tile 120 extends parallel to the direction of conveyer belt 102.In one embodiment, color tile 120 can be perpendicular to conveyer belt Direction slide.In another embodiment, color tile 120 can also slide parallel to the direction of conveyer belt.In some realities Apply in example, relative to the product stream direction on conveyer belt, color tile can either vertically move, and and can enough moves in parallel.One In a little embodiments, color tile 120 may include matt surface, when thus can avoid being illuminated on its surface, in light source or light Occurs dazzle around line.Alternatively, in one embodiment, lighting source may include one along calibration reference unit Length direction reflecting strips or reflector 164.

Lighting unit 160 (showing in the best way in Fig. 1) includes light source 162, and selectively includes reflector 164.Light source 162 can be that product stream that can be to reference calibrations unit and on supply unit 102 carries out appointing for sufficient light What applicable light source.Being applicable some examples of light source includes incandescent lamp (for example, conventional incandescent bulb or halogen tungsten lamp), electroluminescent Light source (for example, light emitting diode (LED)), fluorescence light source (for example, compact fluorescent lamp or plasma lamp), high-intensity discharge lamp, Arc lamp etc..In one embodiment, lighting unit 160 is installed in the side wall of the housing of system 100, as shown in Figure 1.One In individual embodiment, the wall of housing is at an angle of as shown in figure, will not be on imaging object to cause in image acquisition procedures Project shade or dazzle.Due to the product stream on conveyer belt 102 from calibration reference unit be in different height relative to light source On degree, fixed lighting unit may project shade or dazzle in a height, and then will not on another.Reflector 164 can mitigate any shade or glare problem by the steering to light.Reflector 164 is any applicable dress being capable of reflecting light Put, such as mirror or metal surface.The surface of reflector 164 need not polish, because in certain embodiments, it is to " white " Light it is opaque and can reflected white-light it is enough.In other embodiments, for example, when carrying out specific color measuring, can use The selectively reflector 164 of the unwanted light of absorbing wavelength.In the other embodiment measured to fluorescence, reflector 164 may receive a kind of optical wavelength, and be converted into an optimal wavelength that can be then detected.For aforementioned exemplary, It may need using the color tile 120 for possessing particular community, it will can simulate tested feature.Color tile 120 and anti- The design and processing of emitter 164 can use all changes as be known in the art for being used for property needed for incorporation presentation in the substrate Learn the semiconductor type technology of material.

In certain embodiments, reflector 164 and/or whole lighting unit 160 are manipulated by control unit.Should Pay attention to, difference in height the problem of also causing light-intensity difference between conveyer belt 102 and calibration reference unit.It is basic in light source 162 In the case of irradiated object, the luminous intensity that calibration reference unit receives is higher than the calibration reference list on conveyer belt 102 The product stream of first lower section.Thus, in certain embodiments, control unit can be to be in calibration mode or inspection according to system 100 Survey pattern adjusts the output intensity of light source 162.In other embodiments, can be according to intensity of illumination between the distance away from light source Relation calculate and correct intensity of illumination.This relation can be determined by theoretical or experience, be then internalized by calibration algorithm In.

Fig. 3 show the detail of one embodiment of imaging unit.In certain embodiments, imaging unit includes taking the photograph Camera window 150, row scanning camera 300 and control/data wire 350.In other embodiments, the quilt of row scanning camera 300 Any equipment that product stream can be obtained and calibrate the coloured image of reference unit is substituted.Imaging unit is substantially positioned at imaging Thing (for example, conveyer belt 102 and calibration reference unit) top.Camera window 150 by it is any can with it is lossless or with produce Any material (for example, glass or polymer) that the minimum loss because of its own absorption carrys out transmitted light is made.Camera window One of 150 effect be protect row scanning camera 300 and control/data wire 350 from production line surrounding environment influence (such as Dust granule), and efficient cleaning can be realized.

In the specific embodiment shown in Fig. 1 to 3, camera window 150 can extend across the width of conveyer belt 102. In such embodiment, row scanning camera 300 moves across camera window 350, thus obtain object row image (that is, as The linear array of element).Row scanning camera 300 swings back and forth across camera window 350 in a generally continuous manner, to produce One two dimensional image (that is, the accumulation of pixel linear array).In other embodiments, row scanning camera 300 is not swung, but It is rendered as the linear array of neighboring sensor elements sampled to independent pixel.In one embodiment, 2,048 individually Sensor element is arranged using linear pattern, while pixel 0 to 2047 is sampled respectively.This sampling can continuously repeat into OK, to form a continuous two dimensional image.In one embodiment, this two dimensional image be divided into substantially 4 1024 × The square frame of 1024 pixels, each square frame be used to analyze.The size of frame is not limited to any particular value, and can be according to system Demand be adjusted.For example, in certain embodiments, frame can be small to 2048 × 1 pixels, can also arrive only controlled unit greatly The degree that memory used is limited, such as 2048 × 100,000 pixel.View data warp acquired in row scanning camera 300 Data/control line 350 is transmitted to control unit.Data/control line 350 can also drive the movement of row scanning camera 300.One In individual embodiment, row scanning camera 300 is sampled with the speed of about 4000 times per second to all 2048 pixels.This is adopted Sample speed can basically form a square frame under the bottom conveyer speed of about 600 feet per minutes.Obtained using different images In the embodiment of taking equipment, if the visual field width of video camera is enough to obtain image (example on the whole width of conveyer belt 102 Such as, wide angle cameras or video camera), then video camera can be without mobile.In certain embodiments, row scanning camera 300 Equipped with one or more optical sensors, such as charge coupling device (CCD), reverse bias light emitting diode (LED), photosensitive two pole Pipe, complementary metal oxide semiconductor (CMOS), video camera pipe (for example, staticon), or can be by image and input Other any applicable image acquisition equipments for the digital array form being digitized as in the processor of control unit.

Fig. 4 shows one embodiment of the methods described 400 of operating system 100.Method 400 shows calibrating die One embodiment of formula and detection pattern sequence.Herein, " calibration " one word refers to the process of include following aspect:Obtain one Light absorbs data are normalized by the light absorbs data of kind or a variety of reference colors being pre-selected, and one group of formation is final Light absorbs data, detection scan data will be referred to and be compared according to final light absorbs data.It should be noted that " normalization " One word can not exchange with " calibration ", because in certain embodiments, calibration data set and detection scan data group can be by normalizings Change.Being explained further (in the explanation with Fig. 5 A and 5B) in following article, normalization are typically a mathematical procedure, wherein will One group of data is multiplied by a factor, by be allowed to simplify for one it is better simply in the form of, to assist the comparison between data group. In having the scanned embodiment of some reference colors, calibration process includes obtaining the light absorbs data of every kind of reference color, right Every group of data are normalized, and the data group to being finally completed adds up, and detection scan data will contrast this final data Group is referred to and compared.Even in the embodiment using only a reference color, " calibration " one word can not be with " normalizing Change " exchange, because calibration process is also possible that the adjustment (as described below) to light source 162.In light source 162 from calibration scan It is adjusted to detection scanning, to illustrate in the embodiment of the luminous intensity difference in the height lifted, " calibration " one word may also Also comprising the implication for being multiplied by a corresponding proportionality factor.In other words, " normalization " is typically referred to one or more groups of light absorbs Data, based on a proportionality factor drawn inside this group of data, the mathematical operations carried out, and " calibration " typically refers to obtain The overall process for the reference color light absorbs data group that must be finally completed.Embodiment shown in Figure 4, calibration mode generally include Step 404 is to 420 (step 436 is used for recalibration), and detection pattern includes step 422 to 434 (wherein one or more steps To be optional, it is detailed further below).Normalization process can be conceptualized as:Adjust each individually pixel of sensor Gain, allow them in an identical manner respond to illumination light field.For example, positioned at illumination light field edge luminous intensity compared with It is low, because there is no lighting source outside visual field (for example, conveyer belt) edge.This can see in such as Fig. 5 A and 5B example Go out, " left side " side (for example, pixel 0 to 1023 of 2048 × 1 frame of pixels) of a row scanning in one embodiment is shown in figure. Strength level gradually steps up since the at a fairly low level of pixel 0, and its from pixel 512 or so to pixel 1023 then It is basicly stable.Scan (for example, pixel 1024 to 2047) in " right side " side Now quite stable strength level, when close to the right hand edge of scanning, intensity will be reduced gradually.This is referred to as " loaf " effect, Because row scanning looks like the top of a loaf.The mode of this phenomenon of normalization operation processing is:Adjust from edge The gain of camera signal caused by area, to improve signal level, response is set to flatten.When not all light source 162 can essence When the identical energy that really sends or even identical spectral component, may also occur similar effect in visual field.At this In the case of kind, normalization seeks to allow each pixel of image device to have to given excitation even if identical can not also Similar color response.It is slightly different to normalize spectral response of the target on its width and length, therefore, although normalization is Attempt to solve the systematic error in the mode of camera record illumination, but it is only capable of reaching and makes normalization target consistent Effect.In one embodiment, it is the one piece of color uniformly rich beautify-house being secured on a metal caul to normalize target (formica) plate.On the other hand, calibration, which is related to, allows known color (for example, Meng Saier color tiles) through illumination light field, and Color correction table is adjusted, to ensure that the color observed is equal to the normalized color of the ceramic tile.It is referred to as by introducing " panchromatic The ceramic tile (color for representing the whole concern scope of given machine) of many colors in domain " (gamut), in illumination light field Be presented to a given color of video camera, it is therefore possible to ensure that, with sheet during each step the limit forced of change In the precision level being consistent, image is reproduced.

Application of the paragraphs below by introducing system 100 under potato chips quality assurance examination occasion.It should be noted that say below An example in the achievable many versions of alternate embodiment of the bright system 100 for being only the foregoing description.System 100 Digital image analysis can fitted through come any field examined the quality of the production or industry (for example, in the outer of final products See the occasion that (color, shape or size) is considerations) application, including but not limited to food processing, the consumer goods, fashionable dress or vapour Car field.

The method shown in Fig. 4 is returned to, method 400 is since step 402, here, by starting lighting unit 160, to sweeping Region (i.e. imaging object) is retouched to be irradiated.In step 404, system 100 selects in the one group of color for predefining or selecting One color tile 120 with particular calibration color.Calibration color can be appointing in the visible-range of electromagnetic spectrum What wavelength/frequency.In one embodiment, system 100 is furnished with polylith color tile 120, and its number range can be from two pieces always To up to thousands of pieces.Although larger color gamut can be such that system 100 is preferably calibrated to full spectrum, operator (or system) may select or be pre-selected a few color paid high attention in this application occasion.Reference color ceramic tile 120 be pre-selected can be carried out according to the particular demands of examined product, or be carried out for the purpose of covering whole color spectrums It is pre-selected.

In step 406, selected color tile 120 is placed on a scaoffold 104.Color tile 120 passes through cunning Enter in slit 112 or be engaged with, be placed on scaoffold 104.In step 408, by control unit or by operator's hand Dynamic operation, scaoffold 104 are moved into visual field.In step 410, row scanning camera 300 is obtained to color calibration ceramic tile 120 A scanning, and send it to control unit, the latter receives scan data and is simultaneously deposited into memory.In step 412, The intensity and color component for the light that the processor of control unit is absorbed to the sensor in row scanning camera 300 are subject to really It is fixed.

As shown in Figure 5 A and 5B, row scan data can be visualized as a two dimensional plot, and pixel number is (along linear The position of pel array) it is x-axis, luminous intensity is y-axis.Row scan data can be further decomposed as each pixel it is monochromatic into The strength level of point (such as red, green, blue, i.e. RGB compositions).As it was previously stated, it is elevated above conveyer belt 102 in calibration reference unit Position when, it may appear that a problem, because even being that same object is scanned, be absorbed or reflected in each height Luminous intensity also can be variant.Once swept to what grey ceramic tile 120 was carried out in the height of conveyer belt 102 for example, Fig. 5 A are shown The color component intensive analysis retouched;Fig. 5 B then show the same analysis carried out in the height of scaoffold 104.It is clear that Intensity in Fig. 5 B is more much higher than the intensity in Fig. 5 A, has nearly reached the degree for making optical sensor saturation and reaching capacity. When by calibration data compared with the scanning information of product stream, this may result in error.

Thus, the conventional method shown in Fig. 4 is returned, in one embodiment, the strong of color component is determined in control unit After degree, in step 414, system 100 decides whether to adjust light source 162.One adjustment is raising or the intensity for reducing light source 162. In one embodiment, if the intensity measured is higher or lower than a predetermined threshold value, control unit processor can be certainly The intensity of dynamic adjustment light source 162.In another embodiment, control unit processor sends one to the operator of system 100 Prompt (for example, passing through display device), to manually adjust light source 162.And in other embodiments, the operator of system 100 exists In the case of the prompting so done not from control unit, the intensity of light source 162 is adjusted according to scanning result.One In a little embodiments, 414 steps also include the direction of adjustment light source 162.(leaned on when since Fig. 5 A and 5B as can be seen that scanning At nearly pixel 0), the value of intensity curve is very low, gradually stablizes with the continuation of scanning.This result may be derived from Shade or dazzle on object (being in this example grey ceramic tile 120) surface, or " loaf " effect described above.One In a little embodiments, angle or the direction of light source 162 are adjusted by control unit, and in other embodiments, then by the behaviour of system 100 Author carries out these operations manually.After adjustment light source 162 (being probably that adjustment intensity, direction or both adjust), method 400 meeting return to step 410, obtain a new calibration row scanning.It should be noted that 414 steps are optional step, in some embodiments In, this method can proceed to directly to normalization step 416 from step 412.In fact, after normalization step, Ren Hete The shape of fixed intensity curve may all recede into the background.

Step 414 in Fig. 4 is carried out after optionally adjusting to light source, in step 416, the processor pair of control unit Color component intensity curve is normalized.As shown in Figure 6 A and 6B, the initial data from calibration scan (Fig. 6 A) is multiplied by One proportionality factor is (for example, the y of RGB curvesmax), it is more efficiently so as to carry out to draw one group of normalized curve (Fig. 6 B) Compare.Normalised RGB curves are substantially linear, thus can more easily compare the relative intensity of R, G and B component. In addition, normalization figure can more easily be envisioned as the absorption figure line of imaging object (with the reflected light acquired in imaging unit Intensity data it is opposite).When compared with the scanning with product stream, one provided by normalization step 416 is more linear Curve help more easily to detect defect.

After normalization, in step 418, system 100 can selectively decide whether that school should be carried out to additional color It is accurate.If more colors need to calibrate, method 400 is by return to step 404.If there is sufficient amount of color by school Standard, method 400 proceed in next step.In step 420, scaoffold 104 is removed row scanning camera 300 by system 100 Visual field (is carried out) manually by control unit or by operator.Due to each color calibration sequence (for example, step 404 to 420) only part second can be short to, calibration and the progress of recalibration can need not interrupt production procedure.Confirm or calibrate color Processing time may need 1 minute or so, and operating in this time for system is unaffected.If in calibration mode The speed that period conveyer belt 102 moves is not near making product flow attend the meeting greatly the degree of influence quality control, then continuous production Flow is in calibration without stopping.

When exiting calibration mode, system 100 can enter defects detection pattern.In step 422, imaging unit can obtain defeated Send the once row scanning of product on the surface of band 102.As calibration scan, the data gathered pass through control/data wire 350 Control unit is transferred into, and is stored in control unit memory.In step 424, control unit processor can be to defect The color component intensity of detection scanning is normalized.Then decide whether to continue to scan in step 426, control unit processor Product on the surface of conveyer belt 102.If 426 steps are determined as affirming, repeat step 422 and 424.If 426 steps Be determined as negate, processor will step 428 to be stored in the defects of memory detection scanning accumulate.In some realities Apply in example, system 100 will continuously repeat step 422 to 426, while the data transfer obtained not long ago to control unit is stored Device.

In step 430, processor determines to whether there is defect in accumulated scan data.In one embodiment, base It is automatically determined in the defects of algorithm comprising many factors, step 430, described many factors include color, the shape of defect Shape or size.And not all possible defective pixel is equally treated, but be able to may have to every cluster or each group scarce Sunken pixel specifies a performance figure score value.In certain embodiments, the algorithm is given including the defects of being directed to each type The threshold level of the one group of performance figure score value gone out.For example, one group of 300 orange pixel on a given potato chips may It is considered as defect as one group of 10 black picture element.Thus, the threshold level of black defect may be most 15 pixels, and The threshold value of orange defect may be 500 pixels.This point is different from some method of quality control of the prior art, existing In method in technology, any one defect (or characterizing the pixel of defect) of appearance is all treated equally.Threshold value is according to specific The needs of application are selected.Algorithm is stored in control unit memory, and in step 432, processor calls what is preserved Algorithm, to determine flaw size whether higher than related threshold value.If flaw size, higher than related threshold value, system 100 will be to Screening washer sends a signal, to filter out possible defective product, is further examined.If it is not, method 400 will Carry out subsequent step.In step 434 step, system 100 determines whether to continue to scan on the product stream on the surface of conveyer belt 102.If It is determined as affirming, method 400 will continue return to step 422, detection pattern.

It should be noted that the data from imaging unit may be influenceed by the slight change of environment or condition.For example, Even if row scanning camera 300 is focused on color calibration ceramic tile 120, if the color of peripheral region (such as conveyer belt 102) Change, row scanning result can also change.For example, the conveyer belt 102 extended after usage time conveying potato chips may be changed into orange Color is to brown.Some in the light that light source 162 is sent is absorbed by conveyer belt 102.Thus, when will be in clean conveyer belt 102 It is the back of the body for the calibration scan (as shown in Figure 5 B) carried out under background to color tile 120 and in used brown conveyer belt 102 When the calibration scan carried out under scape compares (as shown in Figure 6A), it can be seen that Fig. 6 A blue intensities curve is significant lower.By This, aging and color change with conveyer belt 102, system 100 may need recalibration.In step 436, processor determination is It is no to need recalibration.If it is determined that it is affirmative, method 400 is by return to step 404.If it is determined that being negative, method 400 will be carried out Subsequent step.

Although being described in one embodiment to method 400 above, still had perhaps using the method for system 100 A variety of possible embodiments and version.For example, as described above, adjustment of the step 414 to light source 162 can be one can The step of selecting.Recalibration step 436 can scan after the initial imperfection detection scanning of step 422 or in any subsequent detection Between carry out;After step 432, without carrying out recalibration step 436.In addition, without equal after each and each scanning Step 424 is normalized to detection scan data.The normalization step 424 of defects detection scanning can be in all inspections of step 428 Survey after data are accumulated and carry out.Those of ordinary skill in the art it would be appreciated that, without departing from this disclosure scope Under the conditions of, the orders of many steps in method 400 can adjust, as optional step or basic while carry out.

In addition, system 100 can also change.Specifically, calibration reference unit need not include coloured panel 120. For example, as shown in fig. 7, in one embodiment, reference color is placed on a rotating cylinder 700.Rotating cylinder 700 has more Individual color segment 702,704,706,708,710, can be selected it according to the particular demands of checkout procedure.In one embodiment In, roller 700 stops being placed on one group of pillar 750 of its any end.In one embodiment, control unit rolls rotation Cylinder 700 makees enough rotations, and next color segment is placed in the visual field of video camera.In one embodiment, rotation axis passes through Cross the center of roller cylinder.But color segment and the horizontal bar for needing not be length direction as shown in Figure 7, along rotating cylinder 700 Band.In another embodiment (not shown), color segment is arranged with diagonal way so that is continuously acquired in imaging unit various During the calibration scan of color, rotating cylinder 700 can be rotated continuously.

And In yet another embodiment, color tile 120 is fixed parallel to the direct of travel of conveyer belt 102, as shown in Figure 8. In certain embodiments, parallel coloured panel 800 include color segment 802 that multiple directions parallel to conveyer belt 102 extend, 804、806、808、810.With the traveling of path shown in row scanning camera along dotted line, it will obtain the product on conveyer belt 102 Flow the calibration scan of defects detection scanning and reference color band 800.Thus, each defects detection is scanned, can contrasted With reference to being calibrated.In certain embodiments, there is a barrier in the housing of system 100 by conveyer belt 102 and parallel color Band 800 separates, to prevent from accumulating any dust or waste material on parallel coloured panel 800.In certain embodiments, parallel color Band 800 need not be laid along the whole length of conveyer belt 102, need to only cover the visual field of imaging unit enough.This area For those of ordinary skill, it is to be appreciated that in the embodiment using parallel coloured panel 800, many steps in method 400 can be with It is cancelled or carries out simultaneously.

Although have references to some embodiments in the explanation of the present invention, people in the art person should be proposed that species is unlimited Change, deformation, change, remodeling and change, and be intended to the present invention and cover belong to this in spirit and scope of the appended claims Class change, deformation, change, remodeling and change.By the feature of embodiment being combined, integrating or being omitted the replacement to draw Embodiment falls within the scope of this disclosure.

To assist any reader of United States Patent and Trademark Office (USPTO) and any patent authorized for the application reason Claims appended hereto is solved, it is intended that pointing out, applicant (a) is not intended to require any content to appended claims Section 112 the 6th (6) money of day effective 35U.S.C. in the application submission is quoted, unless in concrete right requirement clearly The words and expressions such as " means being used for ... " or " the step of being used for ... " are used;(b) it is not intended to pass through any statement in this specification, with Any mode not reflected in the following claims limits the invention.

Claims (23)

1. a kind of method for being calibrated to dynamic digital imaging system, the dynamic digital imaging system is used to detect The defects of production stream with more than one piece product on conveyer, wherein methods described includes:
A) top for being increased to conveyer being placed on the calibration color ceramic tile with calibration color in the visual field of imaging unit Platform on;
B) the light absorbs data of the calibration color are obtained using imaging unit;
C) the composition color intensity of the light absorbs data of the calibration color is calculated using processor;
D) by adjust the imaging unit sensor it is each individually pixel gain, to it is described calibration color it is described into Color intensity is divided to be normalized;
E) platform is removed to the visual field of the imaging unit;And
F) the light absorbs data of the product are obtained using the imaging unit.
2. the method as described in claim 1, in addition to:
G) the composition color intensity of the light absorbs data of the product is calculated using the processor;
H) the composition color intensity of the product is normalized;And
I) the performance figure score value of the product is calculated according to the composition color intensity of the calibration color and the product.
3. method as claimed in claim 2, in addition to:
J) call and be stored in the defects of memory threshold data;With
K) by the performance figure score value compared with the defect threshold data.
4. the method as described in claim 1, wherein the platform along parallel to the conveyer by producing the direction flowed Move and be moved out of the visual field of the imaging unit.
5. the method as described in claim 1, wherein the platform along perpendicular to the conveyer by producing the direction flowed Move and be moved out of the visual field of the imaging unit.
6. the method as described in claim 1, include the output intensity of adjustment lighting source.
7. method as claimed in claim 6, wherein the output intensity of the lighting source is to be based on the calibration color Light absorbs data be adjusted.
8. method as claimed in claim 6, wherein the output intensity of the lighting source is the institute based on the product Light absorbs data are stated to be adjusted.
9. the method as described in claim 1, the also direction including the use of reflector adjustment lighting source.
10. method as claimed in claim 9, wherein the direction of the lighting source is the light based on the calibration color Data are absorbed to be adjusted.
11. a kind of calibration system for being calibrated to dynamic digital imaging system, the dynamic digital imaging system is used for The defects of detection production stream, the calibration system includes:
Supply unit, it has the first framework that at least a portion width along the supply unit extends;
Platform, it is promoted to the top of the supply unit by the first bearing, and first bearing is movably coupled to First framework;And
Ceramic tile with reference color, it is located on the platform;
Wherein described platform is configured to supply the gap between the supply unit and the platform;With
Wherein described platform be configured so that when the platform be used for calibrate the dynamic digital imaging system when, the gap It is sufficiently large so that production stream can be by platform lower section without interrupting the continuous of the production stream on the supply unit Flowing.
12. system as claimed in claim 11, wherein the ceramic tile with reference color includes multiple reference color sections.
13. system as claimed in claim 11, in addition to the second framework, straight between second framework and first framework Connect the width across the supply unit, wherein the platform by the second bearing provide further supporting, second bearing with Removably is coupled to the track in second framework.
14. system as claimed in claim 11, wherein the platform is promoted to the upper surface away from the supply unit at least About 7 centimetres of eminences.
15. system as claimed in claim 11, in addition to imaging unit, the imaging unit includes the shooting with visual field Machine, camera window and control/data wire.
16. system as claimed in claim 15, wherein the video camera is the linear array of optical sensor.
17. system as claimed in claim 11, in addition to lighting unit, the lighting unit includes light source and reflector.
18. system as claimed in claim 11, wherein the bearing is the crossbeam operationally moved in orbit.
19. system as claimed in claim 11, in addition to control unit, described control unit include processor, memory and Display device, wherein the system also includes the imaging unit communicated to connect by control/data wire with described control unit.
20. system as claimed in claim 11, wherein the ceramic tile with reference color is included operationally around its longitudinal axis The roller of rotation.
21. system as claimed in claim 11, wherein first bearing is movably coupled to first framework Interior track.
22. system as claimed in claim 11, in addition to:
Slit, it is located in the platform, for receiving the ceramic tile with reference color.
23. method as claimed in claim 2, wherein step h) include each of the sensor by adjusting the imaging unit The gain of independent pixel, the composition color intensity of the product is normalized.
CN201380061353.6A 2012-11-26 2013-11-22 Calibration to the dynamic digital imaging system for detecting the defects of production stream CN104813184B (en)

Priority Applications (5)

Application Number Priority Date Filing Date Title
US201261729883P true 2012-11-26 2012-11-26
US61/729,883 2012-11-26
US13/803,484 2013-03-14
US13/803,484 US9699447B2 (en) 2012-11-26 2013-03-14 Calibration of a dynamic digital imaging system for detecting defects in production stream
PCT/US2013/071490 WO2014082010A1 (en) 2012-11-26 2013-11-22 Calibration of a dynamic digital imaging system for detecting defects in production stream

Publications (2)

Publication Number Publication Date
CN104813184A CN104813184A (en) 2015-07-29
CN104813184B true CN104813184B (en) 2018-02-02

Family

ID=50772946

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201380061353.6A CN104813184B (en) 2012-11-26 2013-11-22 Calibration to the dynamic digital imaging system for detecting the defects of production stream

Country Status (11)

Country Link
US (1) US9699447B2 (en)
EP (1) EP2923217A4 (en)
JP (1) JP2015537219A (en)
KR (1) KR101936457B1 (en)
CN (1) CN104813184B (en)
AU (1) AU2013347859B2 (en)
BR (1) BR112015011930A2 (en)
CA (1) CA2891344C (en)
MX (1) MX345824B (en)
RU (1) RU2636263C2 (en)
WO (1) WO2014082010A1 (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE202016102705U1 (en) * 2016-05-20 2016-06-30 Bobst Mex Sa Quality control station with camera calibration system for sheet element processing machine

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5887702A (en) * 1995-10-27 1999-03-30 Asgco Manufacturing, Inc. Trailing arm secondary belt cleaner
US5991046A (en) * 1998-07-14 1999-11-23 Valmet Automation Inc. Method and apparatus for optically measuring properties of a moving web

Family Cites Families (35)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS501790A (en) * 1973-05-02 1975-01-09
US3840931A (en) 1973-09-26 1974-10-15 Bivens Winchester Corp Drive through car wash with double pivoted spring connected side brush arrangement
US4259020A (en) 1978-10-30 1981-03-31 Genevieve I. Hanscom Automatic calibration control for color grading apparatus
US4454029A (en) 1981-05-27 1984-06-12 Delta Technology Corporation Agricultural product sorting
US4807163A (en) 1985-07-30 1989-02-21 Gibbons Robert D Method and apparatus for digital analysis of multiple component visible fields
JPH04340247A (en) * 1991-03-29 1992-11-26 Nec Corp Reticle-dust inspection apparatus
US5533628A (en) 1992-03-06 1996-07-09 Agri Tech Incorporated Method and apparatus for sorting objects by color including stable color transformation
US5335293A (en) 1992-06-16 1994-08-02 Key Technology, Inc. Product inspection method and apparatus
WO1994028397A1 (en) 1993-05-28 1994-12-08 Axiom Bildverarbeitungssysteme Gmbh An automatic inspection apparatus
US5850472A (en) 1995-09-22 1998-12-15 Color And Appearance Technology, Inc. Colorimetric imaging system for measuring color and appearance
US5818953A (en) 1996-04-17 1998-10-06 Lamb-Weston, Inc. Optical characterization method
EP0920620B1 (en) 1996-08-23 2004-10-13 Her Majesty The Queen in Right of Canada, as represented by The Department of Agriculture and Agri-Food Canada Method and apparatus for using image analysis to determine meat and carcass characteristics
US6011540A (en) 1997-03-28 2000-01-04 Sun Microsystems, Inc. Method and apparatus for generating small, optimized color look-up tables
IL133522D0 (en) * 1997-06-17 2001-04-30 Yuki Engineering System Co Ltd Device for checking sheet packaging
US5963328A (en) * 1997-08-28 1999-10-05 Nissan Motor Co., Ltd. Surface inspecting apparatus
EP1060391B1 (en) 1998-02-20 2003-08-06 The Colorado State University Research Foundation Meat color imaging system for palatability and yield prediction
GB9906765D0 (en) 1999-03-25 1999-05-19 Colour Valid Limited Colour Monitoring
US6577757B1 (en) * 1999-07-28 2003-06-10 Intelligent Reasoning Systems, Inc. System and method for dynamic image recognition
US20010048765A1 (en) 2000-02-29 2001-12-06 Steven Yi Color characterization for inspection of a product having nonuniform color characteristics
US6606403B2 (en) 2000-05-04 2003-08-12 Daniel Freifeld Repetitive inspection system with intelligent tools
US6727452B2 (en) 2002-01-03 2004-04-27 Fmc Technologies, Inc. System and method for removing defects from citrus pulp
JP2004125675A (en) 2002-10-04 2004-04-22 Satake Corp Method for evaluating quality of wash free rice
JP2004157025A (en) * 2002-11-07 2004-06-03 Toppan Printing Co Ltd Multi-sensor head and print quality controlling apparatus using the same
CA2411338C (en) 2002-11-07 2011-05-31 Mcmaster University Method for on-line machine vision measurement, monitoring and control of product features during on-line manufacturing processes
US7195236B2 (en) 2003-03-28 2007-03-27 Northrop Grumman Corporation Automated induction systems and methods for mail and/or other objects
JP2008517285A (en) 2004-10-22 2008-05-22 イーホーフード・アクティーゼルスカブIhfood A/S How to analyze and sort eggs
JP4563847B2 (en) * 2005-03-17 2010-10-13 オリンパス株式会社 Board inspection equipment
US8139896B1 (en) 2005-03-28 2012-03-20 Grandeye, Ltd. Tracking moving objects accurately on a wide-angle video
US7688447B2 (en) 2005-12-29 2010-03-30 Honeywell International Inc. Color sensor
CA2639244C (en) 2007-08-30 2014-10-07 Ossid Llc Labeling apparatus for applying wrap labels and method
FR2929481B1 (en) 2008-03-26 2010-12-24 Ballina Freres De Method and installation of visiometric examination of products in progress
JP5213038B2 (en) 2008-08-28 2013-06-19 独立行政法人農業・食品産業技術総合研究機構 Strawberry quality measuring method and strawberry quality measuring device
US8183859B2 (en) * 2009-06-15 2012-05-22 Carmeuse Lime, Inc. Apparatus and method for calibrating a tramp metal detector
US8284248B2 (en) 2009-08-25 2012-10-09 Frito-Lay North America, Inc. Method for real time detection of defects in a food product
JP5858524B2 (en) 2011-03-29 2016-02-10 ケイミュー株式会社 Method for detecting the paint color of cosmetics

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5887702A (en) * 1995-10-27 1999-03-30 Asgco Manufacturing, Inc. Trailing arm secondary belt cleaner
US5991046A (en) * 1998-07-14 1999-11-23 Valmet Automation Inc. Method and apparatus for optically measuring properties of a moving web

Also Published As

Publication number Publication date
BR112015011930A2 (en) 2017-07-11
MX345824B (en) 2017-02-17
CA2891344C (en) 2019-05-07
RU2015124789A (en) 2017-01-10
EP2923217A1 (en) 2015-09-30
JP2015537219A (en) 2015-12-24
AU2013347859B2 (en) 2017-04-13
CA2891344A1 (en) 2014-05-30
RU2636263C2 (en) 2017-11-21
AU2013347859A1 (en) 2015-06-04
WO2014082010A4 (en) 2014-07-17
CN104813184A (en) 2015-07-29
EP2923217A4 (en) 2016-08-10
KR101936457B1 (en) 2019-01-08
US9699447B2 (en) 2017-07-04
MX2015006649A (en) 2015-12-16
KR20150090158A (en) 2015-08-05
US20140146164A1 (en) 2014-05-29
WO2014082010A1 (en) 2014-05-30

Similar Documents

Publication Publication Date Title
El-Abassy et al. Fast determination of milk fat content using Raman spectroscopy
US9810633B2 (en) Classification of surface features using fluoresence
US8581977B2 (en) Apparatus and method for inspecting labeled containers
US9062859B2 (en) Wafer edge inspection illumination system
US9952151B2 (en) Surface features by azimuthal angle
US7714996B2 (en) Automatic inspection system for flat panel substrate
US5917588A (en) Automated specimen inspection system for and method of distinguishing features or anomalies under either bright field or dark field illumination
US8493558B2 (en) Surface inspection apparatus
TW500919B (en) Surface inspection apparatus
US20160178513A1 (en) Chemical characterization of surface features
CN101021490B (en) Automatic detecting system and method for planar substrate
US9766184B2 (en) Surface features characterization
US9006599B2 (en) Method and device for sorting products
US9863876B2 (en) Reflective surfaces for surface features of an article
TWI289719B (en) Inspection system imager method for inspecting surfaces for defect and method for fabricating electrical circuit
US8836672B2 (en) System and method for improving machine vision in the presence of ambient light
US9201019B2 (en) Article edge inspection
JP6214323B2 (en) Device for distinguishing between dissimilar and unique surface features
US5887077A (en) Method for the recognition and evaluation of defects in reflective surface coatings
JP5825278B2 (en) Defect inspection apparatus and defect inspection method
US9488593B2 (en) Surface features mapping
JP4632564B2 (en) Surface defect inspection equipment
CN1191462C (en) Individual element photoelectric measuring device for plane object
US8373856B2 (en) Optical beam spectrometer with movable lens
KR100850490B1 (en) Inspection method, inspection device, and manufacturing method for display panel

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20180202

Termination date: 20181122